Chromosomal mutations are the underlying cause of most inherited diseases and many developmental abnormalities. Mutations can also lead to alterations in gene expression in somatic cells, leading to loss of the normal differentiated phenotype and ultimately to cellular transformation. We are studying the mechanism of mutagenesis, using an SV40-based shuttle vector as a probe to investigate the molecular mechanisms by which agents that damage DNA induce mutations in mammalian cells. Through use of the shuttle vector, we have extensively characterized the types of mutations that occur in mammalian cells either spontaneously or in response to DNA damage. Analysis of the sequence specificity of these mutations has led to a model which explains how the mammalian DNA polymerase introduces errors during DNA synthesis, causing mutations. Studies with the vector in an in vitro DNA replication system indicate that cellular factors, in addition to DNA polymerase, appear to influence replication fidelity. Further studies with this system should allow a characterization of these factors on the biochemical level. Understanding the mechanisms of regulation of cellular proliferation and differentiation is basic to understanding development of multicellular organisms. One approach to investigating these regulatory mechanisms is to study the behavior of transformed cells. For the past several years, we have been studying mitogenic and antimitogenic growth factors secreted by hamster cells transformed by Ad2 and SV40. Our findings indicate that SV40-transformed cells, but not Ad2- transformed cells, secrete a mitogenic inhibitor (MI) that strongly inhibits a proliferative response in untransformed hamster cells and normal rat cells stimulated with serum mitogens. MI also inhibits a mitogenic response by normal hamster spleen lymphocytes stimulated with lectins that activate T cells (concanavalin A) or B cells (pokeweed mitogen). We have proposed that MI might contribute to the high oncogenicity of the SV40-transformed cells by interfering with mobilization of immune effector cells at the site of tumor growth. We are also using SV40 to study the genetic basis of viral tissue tropism. We find that subcutaneously injected small t-antigen mutants of SV40 often induce abdominal lymphomas in hamsters, rather than the subcutaneous fibrosarcomas induced by wild-type SV40. The mutants may fail to produce a growth factor required for the in vivo transformation of non-proliferating cells.